I'm currently developing a customer survey for our company and one of the questions we are asking is giving me problems.

Right now the question reads:

"Is name of company your first call for your compression needs?" Yes/No
"If no, what would need to be different to make name of company your first call?"

Senior Management is concerned that this question is too negative and wants me to rewrite it. Any suggestions? This survey is my first tasking as the newly appointed Quality Manager and I want to do a good job!

:thanks:

I too am tasked with our quality system and one of the failings we have is our customer satisfaction feedback. I organized a team to develop our customer survey, and we concluded that asking too much will not accomplish anything as the customers will not take the time to fill out the survey.

I developed the questions into a on-line survey for our website and have received 1 response. Not too good. But it was buried and now I have it much more visible and accessible, so our hopes are we get some responses.

Let me ask you. How many customer surveys do YOU fill out. Me, I usually threw them away as I didn't have the time to fill them out. Now, with this quality responsibility, I tend to take time to fill them out.

So I guess you need to look at your survey, put yourself in you customers position and consider if you would take the time to fill out YOUR survey.

If the intent it to get as much feedback as possible, consider WILL YOU get feedback at all?

Make it short and sweet, and make the questions have some value.

Attached is the one our team developed. Pretty basic but it tells us what we need to know. Even with this, we haven't received the feedback I would have hoped for.

By the way, I don't think it is too negative, but question whether anyone will take the time to answer it.

Thanks for the input Oriondad. We actually used a similar survey two years ago and got some really good response from it. The only problem was that since we have multiple customers dealing with multiple departments/divisions, it was difficult to tell where our issues actually were.

I've attached my revised survey with the original question removed and replaced with "What share of your organization's compression needs does Universal Compression supply?"

Any feedback will be appreciated greatly!

Hi, and welcome!

Lucki: May I point out that after the first question, having a "Y/N" in each box for the customer to circle might help? Added later: Ah, I see the 1 - 5 scale above the box. You might want to make it larger/clearer. I estimate you have around 30 seconds per person, at best.

The survey we used was close to what Oriondad has, but I did only have one comment section, in order to keep the survey to just 1 page. I felt that having more than one page would increase the liklihood that it would be tossed immediately.

We both mailed the survey out to targeted individuals and had our Sales Managers hand deliver them to another group. The response rate was > 50%, which is excellent, but the only thing we could really gather from the responses is that for some reason the New England area tended to rate us lower in most of the areas. Of course, this survey was handed out last winter, so maybe that has something to do with it (?).

As I have said before, this clause of 9001, and the interpretation thereof, almost makes me want to disown the whole process. Frankly, I am still looking for a way to get at least SOME value out of it. The fact that our sales call reports are widely read within the company is not enough? Argh!

That said, a search here at the Cove should yield a whole bunch of ideas on the topic. I think you will find, though, that he direction you are heading is pretty close to the consensus here.

Good luck.

Where is a good psychometrician when you need one?

I happen to have friends and acquaintances ranging back into college days in the 60's who have specialized in the creation of polls, tests, and questionnaires designed to get at the "truth."

The fact of the matter is there are many nuances involved in structuring questions to avoid introducing bias in the head of the one writing answers. In over 40 years of experience, I realize this is one of those fields where there is no shame in seeking out a real professional to help you if the answers are important enough to assure they are valid. When I want to create a survey or questionnaire, I often consult with an expert.

Think of it as hiring an attorney when you go to court. You may not bother if the worst penalty you face is a traffic fine. If you are charged with a serious offense or being sued for all your assets, the stakes are higher and you should have an attorney.

If you are going to make decisions and implement actions based on the result of your survey, you should have a professional help you formulate it. If you are not going to make decisions and implement actions, why on earth are you bothering with a survey?

One source of psychometricians who may be available for consulting is a major university in your area. The usual credentials are Ph.D. in Quantitative Psychology, Educational Measurement, or Psychometrics. When you first talk with a psychometrician, he/she should be asking you two questions:

How will you use the information you gather?
Who are the people to be queried (job title, education level, etc.)?
Whether it is preparing an intelligence test, or a certification exam, or a political poll, or a customer satisfaction survey, the psychometrician will try to use scientific methodology to ensure the test questions are fair, accurate, able to be understood by all taking the test (survey), and are precise measures of the test-taker's competency (real opinion) in a given area. The psychometrician also uses scientific reasoning to determine scoring procedures, passing score levels (significance to the organization on whether it should modify procedures), and equivalence among multiple versions of one test.

We both mailed the survey out to targeted individuals and had our Sales Managers hand deliver them to another group. The response rate was > 50%, which is excellent,

We used to do a customer survey as well. Our sales manager would hand deliver them to the customers, and with the biggest 2, he actually sat down an went over the questionaire with the buyer, and got some fantastic feedback. The others that he just dropped off, of course we got no response from. Now we actually measure our customer's satisfaction in-house for the TS required items, and we've redone the survey from a customer satisfaction survey to a customer perception survey, asking some different questions in order to gather an idea of how the customer percieves our company. The Sales Manager will make appointments at our customers, and go over these with the buyers on a one on one basis.

Jonell

I am doing away with mailed surveys, as well as having them handed out by sales reps. I have been given advice that it is better, faster and easier to ask selected customer, new accounts, existing accounts, questions over the phone. I also want to try and develope a set of questions for our sales reps to ask during customer vistis. I tried asking some questions when I was visting an account with our sales rep and explained that I wanted to know how we were doing and then proceeded to ask some questions. The customer was a little surprised that I asked questions that were outside of the typical questions encountered in a sales rep visit. I actualy got some very candidate responses. It was also conveyed that we truely cared about the level of service we provide and that we are striving to do better. I was quite happy with my little field test.
Rich :lmao:

I am doing away with mailed surveys, as well as having them handed out by sales reps. I have been given advice that it is better, faster and easier to ask selected customer, new accounts, existing accounts, questions over the phone. I also want to try and develope a set of questions for our sales reps to ask during customer vistis. I tried asking some questions when I was visting an account with our sales rep and explained that I wanted to know how we were doing and then proceeded to ask some questions. The customer was a little surprised that I asked questions that were outside of the typical questions encountered in a sales rep visit. I actualy got some very candidate responses. It was also conveyed that we truely cared about the level of service we provide and that we are striving to do better. I was quite happy with my little field test.

In this day and age of email viruses, we choose not to email out Customer surveys. Instead we either fax them to our Customers or, for those that frequently order our product online, they are provided with the address to visit to complete an online survey.

We also hand out surveys when they visit us for our Customer Appreciation Days.

And lastly, when our Sales representatives visit their Customers, questions are asked.

Our Customers are just as diverse as our product range and there is no set means for collecting data on their satisfaction level as well as obtaining information on "How can we serve you better?"

Multiple methods allow for a better response rate and, in some cases, more meaninful responses. :)

As is the norm - I have some input on this topic.

We also are doing away with mailing or faxing out a survey. In the past we have surveyed our top 100 customers - which is 97% of our sales. Then within that group we break it down to the top twenty as well because they are 85% of the sales. So we collected data for several years now and have nice charts showing increase satisfaction overall.

However what we found was our response rate (started in 1999 at 84% response) has gone down over the years - Now at 74% - with our biggest customer not responding after several attempts. SO what we are now moving towards is a top 20 program. We take our top 20 customers (which again is around 85% of our total sales) and we schedule face to face time with them to discuss our relationship and their product specifically. At that time we also get the answers to our normal survey to keep the trending of data going.

We feel that focusing on our heavy hitters and giving them personal attention we will increase customer satisfaction with them which will have a greater impact on our business than trying to improve relationships with customers that are .03% of our sales.
While we will not ignore the other customers we are focusing on the top twenty to ensure increased satisfaction, a positive perception and increased growth with them. The other customers will continue to receive our normal achievement and commitment to "High-Quality - On Time Heat Treating".

Hi All,

We send out customer satisfaction questionnaires to our top ten customers on an annual basis (it was initially six-monthly), to get them to rate our performance in key areas and request that they also provide a BIC rating (best in class, the performance of companies supplying comparable goods). Each question has a space in which comments can be entered and a further section is reserved for their comments on any other issues that they may have.

This, in conjunction with return rates, provides us with feedback from our customers and also allows us to compare our performance against that of our competitors.

Phil.

Hi

I became Quality Manager almost 3 years ago and one of the things that I had to do was get our survey program back on track!

We used, and still use, a telephone survey, we actually get one of the employee’s wife’s to make the calls! For the last 3 years we have been getting a response rate up around 70-80%.

Using someone not directly connected with the company is actually very useful as they are independent and can act as a good sounding off board for the customer.

The issue of how to scores surveys is one, which I am interested in. We use the Satisfaction Index %, which seems to work for us. There are, however, some important things to remember with surveys, from my limited experience anyway (I'm still a quality baby!). I would appreciate comments from you guys.

90% of the time, we make a telephone survey for every job we complete. We survey the contact person in the company that we did a particular job for. We have tried mailing surveys in the past but we were lucky if we got 5 or 6 responses.

The problem with this is that it is the satisfaction of that person at the time of the call.

Catch them on a good or bad day and their score may not necessarily reflect that of the company.

However, what it is good for is picking up obvious areas of concern with the last job completed for that person.

Therefore, I am using the system as an indicator of the satisfaction levels but more to focus on the times when we did something wrong.

The other thing to be careful with is the number of samples (responses) you get, especially when you are comparing the satisfaction levels of two different customers or sectors.

Limited responses make it harder to understand what is happening.

It’s all about perception, both the customer and you as the supplier. Most things about customer satisfaction are ‘grey’ and should not be treated as fact.

Hope that helps a little.

Simon

PS If anyone has any useful information on surveys or places to look, I would appreciate it.

Attached is the one our team developed. Pretty basic but it tells us what we need to know. Even with this, we haven't received the feedback I would have hoped for.

By the way, I don't think it is too negative, but question whether anyone will take the time to answer it.


Seems ok to me apart from the opening line stating that you are asking for customer input because QS9000 says that you have to.

Reading it for the first time, much like a customer would, it suggests to me that you are only doing it because you have to and not because you want to?

Also having an central score of average allows customers to take the easy option of selexting the easy ground in the middle, which in the long run may not mean anything to you.

Maybe try to get them to choose one way or the other?

Cheers

Simon

To add to this discussion,important point to keep in mind is the analysis of data.Customer Satisfaction survey data is typically 1 to 5 or 10 scale (Likert Scale). This is not continuous distribution. This is a Discrete data.Hence Mean or Standard deviation is not appropriate.Central tendency is expressed in terms of Median and variability is expressed in inter-Quartile.
Non Parametric statistics is used for Comparison between data sets.
http://www.icbl.hw.ac.uk/ltdi/cookbook/info_likert_scale/

Steve Prevettte in the other forum suggested a very good article"Analyzing Survey Data
On Constructing Interval Scales Using Data Resulting from Categorical Judgements".
http://www.hanford.gov/safety/vpp/survey.htm
I have forwarded this approach to our Corporate Quality folks calculating "Customer satisfaction".

Regards,
Govind.